Anti-EGFR resistant clones decay exponentially after progression: Implications for anti-EGFR rechallenge.

3511 Background: Colorectal cancers (CRC) treated with EGFR inhibitor (EGFRi) evolve by a repetitive process of genetic diversification and clonal evolution. After discontinuation of EGFRi, KRAS/NRAS (RAS) and EGFR resistance mutations no longer have a growth advantage relative to other clones. Yet, the impact of time from the last cycle of EGFRi to decreases in the mutant (MT)/ wildtype (WT) allele ratio has not been well defined in mCRC. Methods: We analyzed a retrospective cohort of patients at MD Anderson Cancer Center with RAS/BRAF/EGFRWT mCRC who had been treated with EGFRi and in whom plasma samples had been collected for sequencing of ctDNA on a platform optimized for very low allele frequencies (AF) (Guardant360). Relative MT allele frequency (rMAF) was defined as ((MT AF) / MT present at the max AF in that assay) and approximates the % of tumor cells shedding the MT of interest into the circulation. rMAF results were normalized to total ctDNA concentration. Results: 135 pts were identified. There was an inverse relationship between the rMAF of RAS and EGFR and time since last treatment, but no change in truncal APC and TP53 inactivating mutations. Our analysis showed that the decline in RAS/EGFR rMAF is best described by an exponential decay model (r = 0.93 for RAS and 0.94 for EGFR). At progression on EGFRi, the median RAS rMAF and EGFR rMAF were 10.5% and 10.6%, respectively, and these clones exponentially decayed with a half-life of 3.4 mo and 6.9 mo, respectively. Our model predicts that at the time of progression only 30% of the cells in the tumor carry a mutation in RAS/EGFR/BRAF/MAP2K1. This suggests that the remaining 70% are driven by undetected autonomous mechanisms of resistance, or paracrine effects. Conclusions: We identified that RAS and EGFR MT alleles decay exponentially over time since last EGFRi. These results provide a molecular explanation for the efficacy of EGFRi rechallenge therapies after a period off EGFR inhibition. The half-life of these clones may help guide timing of rechallenge therapies, and could be monitored by ctDNA. Further, these results suggest that a subclonal tumor population may be exerting resistance through paracrine mechanisms. These data will be validated on an external dataset.